Funding & M&A C

Executives from 01 Bank, a lending-focused digital bank, and Hokkoku Bank discuss how small and midsize firms' fundraising is changing. They argue the decisive factor in lending is shifting from financial statements toward 'data' and 'future scenarios'.

Craig Newmark, founder of the classifieds site Craigslist, has donated half a billion dollars over the years. He is known for philanthropy supporting causes such as journalism, cybersecurity, and military veterans.

Agentic AI systems increasingly rely on language-model components to interpret instructions, exposing them to attacks. This paper analyzes defensive misdirection as a countermeasure against model-guided automated attacks.

The paper presents a process-monitoring approach that combines topological data analysis with machine learning to extract actionable information from high-dimensional time-series. It represents multivariate time-series data for real-time monitoring of dynamic processes.

The paper presents a mechanistic study of representation retention in continual learning, using a controlled toy-world framework to make the drivers of forgetting observable and testable. It examines how sparsity and superposition relate to forgetting, isolating mechanisms that real datasets usually entangle.

The paper introduces CATCH-ME, a dataset of contextually annotated multi-turn counterspeech against overlapping hate speech and misinformation. It addresses NLP's tendency to treat the two threats in isolation and the tendency of zero-shot LLMs to produce repetitive, vague counterspeech.

Large language models have advanced in reasoning, especially deduction. QMFOL benchmarks LLM reasoning through quantifiable monadic first-order logic test-case generation.

Machine learning deployed in evolving streaming environments must handle non-stationarity. This work analyzes and evaluates learner-based approaches to concept drift detection.

The paper presents CzechDocs, a multiway parallel dataset of formatted documents in HTML, DOCX, and PDF covering Czech and minority languages used in Czechia—primarily Ukrainian and English, with smaller amounts of Vietnamese, Russian, and others. It is designed to support evaluation of machine translation.

The paper presents IHUBERT, a monolingual Persian pretrained language model trained from scratch on a RoBERTa-base encoder. It uses vector-based semantic deduplication and domain-balanced pretraining to address the scarcity of large, high-quality Persian corpora and limited evaluation.

OpenAI says GPT-5.5 Instant strengthens ChatGPT's health and wellness responses through better reasoning, richer context, and clearer communication. The work is backed by physician-informed evaluations aimed at delivering more reliable, trustworthy health guidance.

The paper proposes source-grounded data generation for text-to-JSON learning, where models extract information from long unstructured documents into structured, machine-readable JSON. It targets domains such as financial filings and clinical records that store high-value information in unstructured text.

The paper studies generative engine optimization at scale, measuring how brands are represented, cited, and recommended across AI search engines such as ChatGPT, Claude, Perplexity, and Gemini. It frames the shift from traditional SEO as users increasingly get answers directly from these engines.

The paper investigates whether semantic information can be encoded in compact, non-standard text that sacrifices human readability while remaining usable by models. It argues large language models do not always need human-readable language, especially when the intended reader is another model.

The paper presents Prompt, Plan, Extract, a zero-shot agentic LLM workflow for extracting lung pathology information from clinical narrative reports. It targets the labor-intensive, error-prone manual extraction needed for cancer staging and tumor registries, avoiding fully supervised NLP pipelines.

The paper introduces JAMER, a project-level code framework dataset and benchmark for professional game engines. It addresses the lack of large-scale datasets and deterministic evaluation for project-level code engineering, which has remained underexplored despite progress in AI-driven game asset and gameplay generation.

The paper presents CREDENCE, an approach to decomposing compound sentences into atomic, verifiable claims for automated fact-checking. It introduces semantic metrics that avoid token-overlap measures, which underestimate quality for paraphrastic claims, and adds convergence and termination analysis.

The paper presents AgentFinVQA, a deployable multi-agent pipeline for auditable financial chart question answering. It targets regulated settings where practitioners must know which answers to trust and cannot send client data to external model providers, unlike existing accuracy-focused, opaque chart-QA agents.

An article citing leaked financial documents indicating that OpenAI is losing billions of dollars a year. It feeds the debate over the enormous costs of developing and running generative AI and the challenge of turning a profit.

Simulating human users in interactive settings could advance training of agent assistants, evaluation of personalization systems, and social-science research. This work learns user simulators using Turing rewards, aiming to reproduce more realistic user behavior.

Neurosymbolic semantics is fragmented: classical, fuzzy, probabilistic, and neural systems each define truth by their own rules. Extending ULLER, NeSyCat subsumes them under a single inductive definition of truth, delivered as a differentiable tensor implementation for neurosymbolic learning.

AI has driven rapid progress in medical imaging, yielding ever more sophisticated algorithms and steady benchmark gains. Yet this algorithm-centric trajectory reveals limits. This work argues for conceptual innovation beyond algorithms to achieve clinically meaningful advances in medical imaging AI.

Time-series anomaly detection is crucial across applications, and reconstruction-based methods dominate but suffer from over-generalization that reconstructs anomalies too well. SCAN uses multi-scale neighborhood-centered clustering to curb this over-generalization and improve detection.

Family members caring for people with Alzheimer's and related dementias form the foundation of long-term care worldwide; in 2023 over 11 million U.S. relatives provided unpaid care. This work presents a taxonomy of caregivers' mental-health and technology needs to guide supportive design.

Advances in diffusion models and face-swapping enable highly realistic deepfakes and real-world harm. This work shows AUC can mislead when evaluating detectors under domain shift, and proposes a polarization-aware evaluation that better reflects deepfake detector performance across domains.

Clinician-centered evaluation is critical for validating medical AI, especially in ultrasound imaging where quantitative metrics do not always capture clinical usability. This work proposes a clinician-centered pipeline for annotation and evaluation in ultrasound AI studies to ground validation clinically.

The authors introduce Dango, a 1.8B-parameter language model designed for controlled studies of L1-to-L2 (Japanese-to-English) transfer in second language acquisition. By training strictly on L1 only, Dango enables controlled experiments on transfer phenomena that prior SLA model studies could not.

Conversational AI has advanced in language generation, personalization, and long-context interaction, but most methods model social behavior through isolated components. This work offers a formal theory of human-AI coevolution dynamics, explaining how social intelligence emerges through long-term interaction.

Automatic handwritten text recognition is challenging, especially for cursive scripts. This work introduces the Urdu Katib Handwritten Dataset, a historical-document dataset for offline Urdu handwritten text recognition, providing resources to advance recognition research on cursive scripts.

Early detection of cognitive impairment relies on neuropsychological tests whose scoring is subjective. This work mitigates scoring errors and compensates for nonverbal subtests in speech-based dementia assessment, aiming for more objective and reliable screening.